Relevant and even prescient commentary on news, politics and the economy.

A Bit More on Debt and Growth

Here is a bit more analysis of the data set used by Reinhart and Rogoff (R-R) and by Herndon Ash and Pollin (HAP) in their critique, and in particular the stata data set RR-processed.dta with data on public debt to GDP ratios and real GDP growth in 20 developed countries since 1946.  This post is a minor addition to my earlier post and, especially adds little not shown by  Arindrajit Dube.

In general macroeconomists have a problem that we non experimental data so we observe correlation but we want to know about causation.  We don’t know what to do about this problem, so we almost all do the simplest thing which is look at timing — the cause comes before the effect. This is what Dube did showing a much stronger relationship between the public debt to gdp ratio and real GDP growth in the preceding 3 years than between the ratio and growth in the following three years.  Also the non parametric estimate suggest a much stronger relationship between higher debt and lower subsequent growth for low levels of debt (up to 30%  of GDP) than for higher levels — the opposite of what policy makers guessed given the original Reinhart and Rogoff analysis which really didn’t address that issue.

I redo Dube with parametric regressions which have the nice feature that there are standard statistical tests of null hypotheses concerning parametric point estimates (that’s fancy talk for STATA gives me t-statistics).  The very first standard way to look at causality using post hoc ergo propter hoc is caused a Grander causality test. It is just a regression of one variable on lagged values of the explanatory variable and lagged values of the dependent variable.  This is definitely the first thing macroeconomists do does if he she or they wonder about the direction of causality.

Recall the basic regression which notes that a debt to GDP ratio 1% higher is associated with a real GDP growth rate which is 0.02% lower (this is in fact a big deal).



dRGDP is the rate of growth of real GDP in the country, debtgdp is the ratio of public debt to gdp.

Now  l1drgdp is the rate of real GDP growth lagged one year.


Including that variable appears to show a striking decline in the coefficient on the debt to GDP ratio,  but really one should calculate the long term effect of a permanent increase in the debt ratio using

(1-0.3793671)(new steady state growth rate) = constant -0.089041(debt to GDP ratio) so the estimated long term effect is about -0.015 not much smaller than the simple coefficient.

The t-statistic is calculated assuming that the disturbance to growth is independent across countries (no global recessions or oil shocks) and that all structure within a country is captured by the simple lagged term.  STATA is very willing to recalculate standard errors assuming, for example, that disturbances in the same year may be correlated.
. reg dRGDP debtgdp l1drgdp, cluster(Year)

gives a t-statistic on the debt to GDP ratio of -2.45 (the point estimates are exactly the same the cluster option just calculates more robust estimates of the standard errors).  No big change.

Now I can get the t-statistic to be insignificant (you can always do this).  I do this by definining a new lagged variable al3drgdp which is the average of the growth rate of real GDP lagged one, two, and three years.   When I toss that in the regression and calculate standard errors allowing correlation in the same year across countries I get the t-statistic insignificant.


There is nothing wrong with this regression, except that I fiddled till I got the absolute value of the t-statistic under one (the added variable is due to Dube so there is that).    The point estimate of the effect on growth isn’t much lower.  the long run guess is a bit below 0.01 so slightly less than half the estimate based on the simple regression.

There is nothing much to see here.  Neither strong evidence that the original estimate is due to reverse causation nor strong evidence that high debt causes low growth.

Comments (9) | |

World Industrial Production

I see some evidence that higher domestic energy production and improved competitiveness of US manufacturing is having a positive impact on the US trade balance and US industrial production. But I still expects the impact to be limited. In particular, the real non-petroleum trade balance continues to deteriorate. Increased supplies of cheap natural gas helps the petrol-chemical industry, but  its impact on other industries appear limited. If their is any segment of the economy where the rule of one world price holds, it is energy. Cheap energy prices in the US will only last until the industry can remove the bottle-necks and open those markets to competition. Firms do not build new plants to exploit temporary advantages.  Work is already underway to reverse a pipeline from the Gulf Coast to Cushing, Oklahoma so the surplus oil in the Midwest can be exported.  The Boston  Globe reported that an application was just filed to reverse the oil pipeline from the Maine Coast to Montreal to facilitate exports of Canadian tar sands oil.

If you compare US industrial production to the rest of the world there does not seem to be much of a change that can not be traced to recent world economic growth rather than structural changes in US production.


Clipboard01 world a



Clipboard01 world b

In real terms the total trade balance is improving because of reduced oil imports and expanded oil exports.

But, excluding oil, the real trade balance is still growing.  The rate of growth may be slowing, but that seems to stem more from weak  world growth  rather than structural changes in the US.


Clipboard01 trate bal

Moreover, real non-petroleum exports are clearly peaking and may be rolling over.


Clipboard01 exports


Tags: Comments (0) | |

Oh hail the Great Cactus

Looking for some back articles and wanting to try out posting with our new setup I found this one by Mike Kimel (formally known as Cactus) from July, 2010.   It is always fun to look back.*

He presents his data and argument regarding where we’re going and concludes:

This time, I’m not as comfortable; given where and how the Fed has been putting Money I just don’t see increases in the real money supply as being quite as effective as normal. The money is going to fill in a big hole the financial industry created in its collective balance sheet, and isn’t necessarily leading to a lot of additional spending. Furthermore, with all the talk of austerity, it wouldn’t be surprising if the Federal Government starts cutting back on spending.

Given that the weight of the evidence seems almost equally balanced on both sides, this little thing tips it slightly for me: unless and until the Fed starts removing money from the system, I don’t think we’re going into a second dip. But given the Federal Government’s current policies, I don’t expect much more than mediocre growth for the next few quarters either.

Hey Cactus, you thinking of doing a hedge fund by any chance? : )

*Read the post and then the comments to get the greatest fun factor effect.

Tags: , Comments (0) | |

Miranda Rights 101 and Enemy-Combatant Law 102. No, Make That 345 and 346 (Advanced Seminars).

The two immediate what-everyone’s-talking-about legal issues in the Dzhokhar Tsarnaev case concern his Miranda rights–that is, at what point must he be read his Miranda rights notifying him that he has the right to remain silent and to the counsel of an attorney–and whether he can, and if so should, be classified as an enemy combatant under post-9/11 laws.  Dan Crawford has asked me to post on the Miranda issue, and so I will, along with the somewhat overlapping but distinct enemy-combatant issue, but with the caveat that I have no great expertise in either Miranda-rights law or enemy-combatant law.

These issues concern three provisions in the multi-guarantee Fifth Amendment.  The Amendment reads:

No person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a Grand Jury, except in cases arising in the land or naval forces, or in the Militia, when in actual service in time of War or public danger; nor shall any person be subject for the same offense to be twice put in jeopardy of life or limb; nor shall be compelled in any criminal case to be a witness against himself, nor be deprived of life, liberty, or property, without due process of law; nor shall private property be taken for public use, without just compensation.

The Miranda issue concerns the self-incrimination clause, the clause that reads “nor shall be compelled in any criminal case to be a witness against himself.”  The enemy-combatant issue stems from the lengthy first clause, the one that requires presentment or indictment of a Grand Jury in capital or other infamous crime (now interpreted as all serious crimes, whether or not they’re infamous, although this crime certainly passes the infamy test, so the originalists and textualists should be happy), and also from the due process clause.

Okay, and also from the insistence of certain Republican Senate grandstanders that Tsarnaev be held as an enemy combatant.  At least if issue is an operative word in enemy combatant issue.  Which of course it is.

I’ve read several good news and blog analyses discussing these issues in the last couple of days, but the most comprehensive one on Miranda is a blog post by Orin Kerr (h/t Bill H), a law prof at Georgetown and a former clerk to Justice Kennedy, at The Volokh Conspiracy, a libertarian/right-leaning blog where he writes regularly.  (Kerr is the least right-wing of the several writers there, all of whom are law profs, most of them also former Supreme Court law clerks.)

The essence of Kerr’s post–and a key point made also by Katy Waldman at Slate, in another good article on the subject–is that under Supreme Court jurisprudence dating back about three decades, the issue of the constitutionality of a failure to give a Miranda warning arises only if, and then only when, the prosecution attempts to use the defendant’s pre-Miranda-warning statements at the trial.  Obvious examples are a confession, an acknowledgment that the defendant knows another of the defendants, an admission that the defendant was at a particular location at a particular time, or that the gun used in the crime belongs to him.

It is in some respects–and until now I had thought of it as, in essence–part and parcel of the more generic “exclusionary rule,” which prohibits prosecutors from using evidence at trial that was obtained unconstitutionally.  Usually (but, I believe, not exclusively), the formal exclusionary rule comes into play when evidence is seized in a search that violates the Fourth Amendment’s search-and-seizure provision–a warrantless search that does not fall within the Supreme Court’s seemingly-metastasising exceptions to that constitutional provision (freedom! liberty! originalism! textualism!)

The exclusionary rule was developed by the Supreme Court to effectuate the Fourth Amendment, rather than the Fifth Amendment, but to the extent that I had thought about them at all (which is somewhat, but mainly after a new Supreme Court case on one or the other was issued), I still had thought of them as essentially the same.  But Kerr and Waldman point out that under Supreme Court Miranda jurisprudence, prosecutors can use some tangible “fruits” of an improperly un-Mirandized statement; just not the statement itself; the prosecutor can use the gun that the defendant told them where to find, but the prosecutor cannot tell the jury that the defendant told them where he put the gun.

A more important distinction, though, both in the Tsarnaev case and, well, for you and me, is this: Law enforcement interrogators can ask the defendant questions during an improperly un-Mirandized interrogation but then cannot use the defendant’s statements at trial.  But law enforcement cannot just force their way into your home, or search your car, or search you, in violation of the Fourth Amendment provided that they don’t later try to use what they found as evidence at trial or try after the unconstitutional search to get a search warrant based on the what they found during the unlawful search.

That’s because there is a difference between the very nature of the Fifth Amendment self-incrimination provision and the nature of the Fourth Amendment search provision, according to the Supreme Court.  The Court has interpreted the former as a bar to compelled self-incriminating trial testimony–against being a trial witness against yourself. The Fourth Amendment search clause protects against the actual search, independent of your rights at trial.

I think that’s a distinction that some commentators are missing in the Tsarnaev case, and in light of the unusual specifics of this case, it strikes me as as pretty important.  As in, calm down, fellow civil libertarians.  For now, anyway.

The purposes of most law enforcement interrogations are to try to solve the crime, to obtain enough evidence to gain a conviction, and (often, as part of solving the full crime, including learning its breadth) to identify others who participated in the crime.  In this case, though, there is unequivocally no additional evidence necessary to successfully prosecute Tsarnaev.  They even have the statement from the owner of the hijacked Mercedes SUV that the brothers confessed to him that they were the Marathon bombers–unnecessary icing on a very large, multi-ingredient cake.  They do not need a confession, nor any lesser acknowledgement or admission from Tsarnaev, in order to successfully prosecute him.  Nor to argue for the death penalty.

What they do need is to know with certainty that there are no explosives still stored somewhere, and that there are no other members of their terrorism conspiracy.  Both appear unlikely, it certainly seems.  They probably took all their explosives and guns with them on their wild ride late Thursday and early Friday (they had a lot with them).  The older brother reportedly was strongly disliked among the members of the Cambridge mosque he attended, and Muslim Causas separatists are, reportedly, just that: separatists at war with Russia.  So if the older brother learned his explosive-making craft overseas, it is almost certain that he wasn’t enlisted there to explode bombs at the Boston Marathon.  But law enforcement does need to try to set these issues to rest.

What law enforcement does not need is to try to use un-Mirandized statements by Tsarnaev at his trial.  If the Department of Justice does try, that would be an absolutely unnecessary attempt to distort and stretch Miranda jurisprudence, and the only conceivable purpose would be a decision by Obama (who presumably will be making politically-charged calls in the prosecution) to wave a red flag in the face of civil libertarians, as part of his ever-present quest to be viewed as a “centrist,” and the failure he shares with so many political pundits and other pols to recognize that 2002–like 2010–has passed.

The Marathon bombings are not 9/11.  They are instead the Oklahoma City Federal Building bombing, in which the two perpetrators held radical ideological views that they shared with many others, including members of large loosely-connected groups that advocate violence and that hold deep grudges against the American government.  But those groups, in this case including al Qaeda, likely were not a part of the acts of terrorism perpetrated by the two pairs of perpetrators.

Which brings me to the issue of enemy-combatant status, and of Lindsey Graham et al., who themselves think it’s still 2002 or, more likely, think voters do.  But as Rand Paul can attest, the public doesn’t.  Including much of the Tea Party public.  Even maybe in South Carolina, where Graham hopes to fend off a Tea Party primary challenger.

There are by now enough articles and blog posts published since Friday night, by people with extensive knowledge about enemy-combatant law–which I certainly do not have–that I think I should just say, in summary on this, that, according to those articles, the law does not permit Tsarnaev, a United States citizen charged with committing a crime on United States soil and apprehended not on an overseas battlefield but instead in Watertown, MA, from being declared an enemy combatant.

But to those who insist otherwise, I suggest that they beware of the double-edged sword that they want to manufacture.  Over the weekend, some Republican senator (I don’t remember which one) pronounced the United States “the battlefield”–terminology used by the Supreme Court in their post-9/11 enemy-combatant opinions–because the act terrorism occurred in the United States.  Any act of terrorism in the United States, he claimed, renders the United States a battlefield.

Which in turn requires a definition of “terrorism,” I would think.  And since there is no evidence that the Tsarnaev brothers were connected with any foreign group at all, much less one in which the enemy-combatant statute applies, this senator is proposing, if unwittingly, that anyone accused of, say, using a semi-automatic weapon loaded with a huge magazine could, and should, be declared an enemy combatant.  A designation whose purpose is to strip the defendant of constitutional due process rights and allow permanent detention without trial.

I say, bring it on.  Gitmo for anyone suspected of using a semi-automatic assault rifle with large magazines in the commission of crime!  That’ll have to do in lieu of Gitmo for silly, hypocritical politicians.

Tags: , , , , , Comments (10) | |

Full Cred and Props to Reinhart & Rogoff and the BEA: They Collected the Data

The other day I dissed the analysis in Reinhart and Rogoff’s Growth in a Time of Debt as being on the level of a blog post from an amateur internet econocrank. I still hold that opinion.

But I want to walk back on that, or at least clarify, and give lots of credit where due. Because they did make a huge contribution, of a quality that you will not find in econoblog posts from even the best bloggers: they assembled a great data set. As I can attest — having spent hundreds of hours assembling data sets that were far less challenging than theirs — this is not a trivial task. And the value of that data set is high, assuming you throw high-quality analysis at it.

Now of course, they didn’t release the goddam data set for years. That seems unforgivable, especially given the paper’s political and policy impact. This paper wasn’t in a peer-reviewed journal (it was a “discussion paper,” which makes its impact even more eyebrow-raising), but I really wonder why those journals (in any field) would publish such papers without requiring that the data sets accompany them, for vetting by other researchers. (Yeah I know: not a new idea.)

I totally understand why this is problematic. This is the researchers’ crown jewels, upon which they can build future papers, at least. Not just the data, but the analysis methods and the coding of those methods (intellectual property?), is often included in the files. At most, you’re looking at corporate/university/personal assets, trade secrets, generally some (at least potentially) damned valuable stuff. (Throw in the issue of partial or complete government funding for the research, and it’s even more complicated.) But providing the data should be the default requirement, with some clear guidelines justifying and explaining why the data is not provided, when it isn’t.

The “damn valuable stuff” double-points to the other topic I want to mention here: the BEA’s move change the NIPAs, to count spending on R&D and the development of creative works as investment spending rather than consumption spending — as real-capital building.

“Double” because 1. the new accounting highlights the real value of this kind of data-gathering and knowledge-creation, and 2. the change itself required (and will continue to require) a huge amount of data gathering.

I was kind of wowed by this line in the Financial Times writeup on the change:

The Internet Movie Database may not seem like a natural source of data for the national accounts, but it was one of many combed by BEA researcher Rachel Soloveichik, who went through film studio records as far back as the 1920s to build a series on investment in movies.

(Another good FT post on this here.)

So when you hear me kibbitz about the structures and methods used in the national accounts, please know that I have wide-eyed respect for the diligence and skill of hundreds of accountants and economists involved in building those structures, and populating them with data from hundreds of diverse sources. (This actually sounds like one of those Google interview/hiring questions: how would you go about estimating the value of every movie made in America since the 1920s? Books? TV programs?)

That information is hugely valuable. It’s a great example of Your Tax Dollars at Work, delivering value far above the government’s cost.

Or at least, I think it is. Somebody should gather some data on that.

Cross-posted at Asymptosis.

Comments (9) | |

Due Process: Holder vs Colbert. Art or Reality. Choose.

Re-posted from last year is Dan Becker’s post:

Due Process: Holder vs Colbert. Art or Reality. Choose.
This is the object:

No person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a Grand Jury,… nor be deprived of life, liberty, or property, without due process of law;

It’s all one sentence. Any questions?

This is Art:

The lyrics of Grand Funk Railroad’s Paranoid

Did you ever have that feeling in your life

Tags: , , , , , , Comments (2) | |

Identity Games: Saving ≠ Saving? Whodathunkit?

I finally figured out a simple way to explain my confusion (and that of many others, including many economists) with the whole Saving issue. I may also have figured out a useful solution to that confusion, which I present at the bottom here for my gentle readers’ delectation and denunciation.

Econ profs: I’m really curious. Do you think this post would help your intro students understand this stuff?

First: The accounting’s fine. Of course. But for some not-crazy reasons, the definition of “Saving” changes in the course of the accounting.

Thinking of the “real” sector for the moment, for simplicity and clarity. For each of the economic units at the bottom level of that sector (households and nonfinancial businesses), Saving means money saving:

(1) Saving = Income – Expenditure

But at the top, the level of “sectoral” saving, Saving means saving of real goods:

(2) Saving = Income – Consumption Expenditures

Or in words that more aptly describe what’s being depicted:

(3) Saving = Production – Consumption

(Reminder: Consumption Spending + Investment Spending = Expenditures = Income = Production)

Explanatory aside: There’s Gross or Net Saving, depending on whether Consumption just includes Consumption Spending (on goods that are bought and consumed within the period), or also includes Consumption of Fixed Assets — the very real “depreciation” of those assets. Gross is long-lived goods produced; Net is long-lived goods added, above and beyond what’s “consumed.”

Back to identities: Unlike every other measure in the national accounts, if you sum up the money Saving of all the bottom-level units, it doesn’t equal Saving for the sector. Rather:

(4) Sectoral Saving = Units’ Combined Money Saving + Investment Spending*

Investment spending, of course, causes the creation of real, long-lived goods. But this is the thing that has confused me from the get-go: Saving is (savings are) some combination of money and real goods? Aren’t financial assets supposed to be representative of, proxies for, the real assets? (Equally confusing: economists’ insistence on talking about “capital” as if it were some undifferentiated, homogeneous or vaguely contiguous lump of real and financial capital.)

Here’s what you need to know to sort that out: You know that money saving? It’s zero.

Comments (13) | |

Seven Steps to Social Security cuts via Debt Commission…following the plan

This is a re-posting of a Bruce Webb post from January, 2010.  A reminder of a well crafted plan to date:

Seven Steps to Social Security cuts via Debt Commission

For those of you who didn’t go the short version of my comment is that the strategy to get major slashes to Social Security and Medicare takes an Seven Step and that this technique is not new and in fact mirrors the original plan for Bush’s Commission to Strengthen Social Security (CSSS) in 2001-2002.

Step one:  Get consensus on ‘Crisis’. In this case that current debt growth levels are unsustainable.

Step two:  Get consensus that there are only three possible paths out: revenue increases (A), cuts in military and other discretionary spending (B1 and B2), or cuts to non-discretionary spending, meaning Medicare and Social Security (C)

Step three:  Having agreed that some combination of A, B, and C is needed set up a Commission with a mandate to propose an up or down vote.

Step four:  Committee decides it is unwise to increase taxes during a recession and eliminates (A). Commission further decides that it is unwise to cut defense spending in the middle of two wars eliminating (B1) and that eliminating infrastructure spending or farm supports is both unwise or politically impossible in the current climate (B2)

Step five:  Recommend a package of cuts to Medicare and Social Security-C on the basis of shared sacrifice, after all every CD and State has a share of the elderly population.


Step Six:  Tell Congress that the statute doesn’t allow them to revisit A or B and then that NOT voting for a C based solution means denial of Steps one and two.

Step Seven:  Either get a vote for C or run against opponents as ‘Do Nothing Deficit Deniers”

This was exactly the plan used by the Bush Administration in their Spring 2005 Social Security Tour. Rather than produce a Plan already in hand crafted to specifications, say like Model 2 of CSSS, and have people pick it to pieces, they held it back for a later stage. . So the proposed sequence there was:

Step A: Use Social Security Tour to get consensus on Social Security Crisis
Step B: Use tour to assure everyone that all options were on the table.
Step C: Having gotten consensus from Congress try to find some mechanism to assure a final up or down vote.
Step D: Having secured a vote indicate that any plan MUST comply with the seven existing guidelines of CSSS Guiding Principles to CSSS which bar any tax-based solution and mandate private accounts.
Step E: Remind Congress that they promised an up or down vote and that refusal just meant being in denial of what was conceded in step one and two.
Step F: Either get a vote to ‘reform’ Social Security or use a denial to go for a major victory in the 2006 mid-terms.

In 2005 Bush mostly got stopped at Step A. Since his proposal was narrowly focused on Social Security, a ‘There is No Crisis’ narrative WITHIN the Social Security context was able to get traction. This time we are on a very different track, instead of selling this as a proposal for an ‘Ownership Society’ (where the numbers were pretty easily debunked), it is being sold as a matter of ‘Intergenerational Equity’ and ‘Fiscal Responsibility’. And from that starting point Step one is pretty much in the bag and logic gets you mostly through Step two.

Leaving us where? Well we are within a week of a vote on Step three of the first list and if it passes steps four through seven pretty much follow automatically, they may not work but I would hate to have to bet on it. Meaning that we need to stop Step three by convincing people that Step four is already in the bag. So called ‘Deficit Hawks’ strongly overlap with ‘Tax Hawks’ and with ‘Military Hawks’. Moreover they are largely from farm states and not likely to vote for big cuts there. Which really leaves only one question in my mind, do they stop with proposing big slashes to Social Security, Medicare and Medicaid? Or make new runs at Urban Transit, Community Develpment, or (non-military) Foreign Aid?

I suspect they know better than to get too ambitious and will instead just strike at Entitlements as such.

A last note on Bartlett who revealed too much. In noting that while open to tax increases in principle history showed that while Congress couldn’t help cutting future tax increases back (as with AMT), both Congress and people allowed the benefit cuts in the 1983 Reform to occur on schedule. Since future wage working retirees actually decided to go along with that for the general good while the wealthy would predictably resist paying taxes for that same general welfare that we should just go with the benefit cuts. Because they were ‘doable’.

So workers are ‘doable’. Which puts us ‘working guys’ into a whole new category, except in this case we are paying the Johns.

Tags: Comments (3) | |

A Bit on Public Debt GDP Growth and Causation

Here I  analyze the data set used by Reinhart and Rogoff (R-R) and by Herndon Ash and Pollin (HAP) in their critique and in particular the stata data set RR-processed.dta with data on public debt to GDP ratios and real GDP growth in 20 developed countries since 1946.

I show evidence that low growth causes a high debt to GDP ratio, so the correlation can’t be interpreted simply as the effect of debt on growth (damn fractions how do they work anyway). Of course this has been shown much better and much more thoroughly by HAP at PERI (warning the sudden traffic killed their hamster) and then their colleague  Arindrajit Dube using non parametric regressions.

The following paragraph is pointless and may be skipped. <pointless nerdo twittosity>My added bit is a bit nerdo twitty — I note you can test whether the debt to GDP ratio is a good regressor or instrument (technically is it weakly exogenous) by considering it and another variable as instruments for it.  The other variable should not affect growth except through the debt to GDP ratio.  This motivates a regression on the debt to GDP ratio and the 5 year lagged debt to GDP ratio.</pointless nerdo twittosity>

This is a regression of one years real GDP growth on the debt to GDP ratio and the 5 year lagged debt to GDP ratio ( l5debtgdp)

OK that’s about it.   If one trusts the standard error calculation, one concludes that there is very very strong evidence that, given debt now, it is much better to have been highly indebted already 5 years ago.  This is the pattern one would expect if low growth caused a high debt to GDP ratio.  Future growth is low if debt is higher than one would predict given debt 5 years ago — presumably because that is the result of disappointing growth and growth rates are serially correlated.  Old debt is not so damaging.  This means that it comes out looking as if old debt is positively a good thing (really the regression doesn’t show this it shows if you have debt it is better for it to be old).

Ooops I motivate the regression by assuming that growth rates are correlated over time, but the standard errors are calculated assuming they aren’t. Fortunately STATA makes it easy to  correct standard errors for correlation of any kind within groups of data — in this case growth rates for the same country at different times.

Just type

. reg dRGDP debtgdp l5debtgdp  if l5debtgdp!=.,cluster(cntry)

gives a new T-statistic on l5debtdgp of 3.19 markedly smaller than the uncorrected t-stat of 5.11 but still very significant.

The regression basically rejects the null (never stated but often insinuated by R-R) that the pooled OLS regression of read GDP growth rate on the debt to GDP ratio gives a valid estimate of a structural causal relationship.

Many more regressions and the batch program I used are here

Comments (2) | |